Produces a consultant-grade machine-guarding assessment for a named machine or line — guard-by-hazard-zone per ISO 12100/14120 and OSHA 1910 Subpart O (1910.212/.219). Use this skill whenever a user asks to assess machine guarding, review a guarding survey, select guards or protective devices for a danger zone, check guarding against PUWER or Subpart O, or build a hazard-zone register — not a generic 'guard all moving parts'. It is engineering-control-led: each danger zone walks the selection order (fixed, interlocked, presence-sensing, two-hand, trip) against the access-frequency rule, and a mechanical-zone control left as PPE-only or admin-only ('keep hands clear / wear gloves') is flagged and pushed up the hierarchy. It refuses 'a machine' or 'looks guarded' without a named machine plus motion/hazard type plus safeguarding status, cross-references the existing lockout/tagout map for maintenance, and emits a branded hazard-zone register. Decision-support only; a competent person must review the output.
Produces a consultant-grade machine-guarding assessment for a named machine or line — guard-by-hazard-zone per ISO 12100/14120 and OSHA 1910 Subpart O (1910.212/.219). Use this skill whenever a user asks to assess machine guarding, review a guarding survey, select guards or protective devices for a danger zone, check guarding against PUWER or Subpart O, or build a hazard-zone register — not a generic 'guard all moving parts'. It is engineering-control-led: each danger zone walks the selection order (fixed, interlocked, presence-sensing, two-hand, trip) against the access-frequency rule, and a mechanical-zone control left as PPE-only or admin-only ('keep hands clear / wear gloves') is flagged and pushed up the hierarchy. It refuses 'a machine' or 'looks guarded' without a named machine plus motion/hazard type plus safeguarding status, cross-references the existing lockout/tagout map for maintenance, and emits a branded hazard-zone register. Decision-support only; a competent person must review the output.
A consultant-grade, guard-by-hazard-zone machine-guarding assessment for a named machine,
line, or cell — never a generic "guard all moving parts" statement. Its entire reason to exist
is that mechanical-zone safeguarding is engineering-control-led: for every danger zone the
guard/device is selected in order — — against the , inside the
(inherent safe design → safeguarding → information for use). A mechanical-zone control left as
("operators trained to keep hands clear / wear gloves" with no fixed or
interlocked guard) is . That is the failure mode this skill exists to prevent.
flagged and pushed up the hierarchy — never accepted as the headline
control
It forces the single lever that separates a defensible artifact from copy-paste paperwork:
per-zone specificity plus the full hierarchy of controls. It refuses to assess "a machine"
or "looks guarded" without a named machine + the motion/hazard type + the existing-safeguarding
status. The maintenance interaction mode cross-references the existing KB-REG-LOTO
energy-isolation map (lockout/tagout). Grounded in OSHA 29 CFR 1910 Subpart O (1910.212 /
1910.219), ISO 12100 / ISO 14120, and PUWER 1998 Regs 11–12. Decision-support only; a
competent person must review the output.
When to use this skill
Use this skill when the user needs a machine-guarding assessment for a concrete machine, line,
or cell — for example "assess the guarding on the PL-3 power press", "review the guarding survey
for the assembly cell against Subpart O", "select guards for the in-running nip point on the
calender", or "build a hazard-zone register for the CNC mill". It is not for a generic
"how do I guard machines?" answer: the Workflow intake below forces the named machine, the motion
and danger zones, and the existing-safeguarding status before any drafting, and refuses a vague
"a machine" / "looks guarded" request.
Data Protection & De-identification (MANDATORY — apply before drafting)
Apply this BEFORE you draft anything. Treat injury, illness, and any health
detail as the highest sensitivity. Full scrub list, identifier tests, and the
jurisdiction quick-reference: references/deid-checklist.md.
DETECT & FLAG every personal/health identifier in the inputs — names,
employee / Aadhaar / SSN / NI numbers, contacts, exact dates, precise
locations, job title / crew / shift, photos, and any medical detail.
List what you found before drafting. If unsure whether something is
identifying, treat it as identifying.
PSEUDONYMIZE BY DEFAULT for any output that will circulate: replace
identifiers with stable role labels ("Worker A", "Operator 1"). Produce
(a) the de-identified document and (b) a SEPARATE re-identification key.
Never put the key or any name↔label mapping in the document. Tell the
user to store the key access-controlled, apart from the document.
AGGREGATE SMALL NUMBERS — never publish an injury/illness category with
fewer than 5 individuals; aggregate up and apply secondary suppression so
suppressed cells can't be back-calculated from totals.
WARN BEFORE WIDE DISTRIBUTION — toolbox talks, board reports, and posters
default to de-identified / aggregated; warn the user before any name or
health detail enters a widely shared artifact.
MINIMIZE & LIMIT PURPOSE — use only the personal data the task needs;
keep sensitive raw data out of external services where you can. When in
doubt, ask before including it.
Knowledge base (read ONE matching file — never load all)
Resolve the user's jurisdiction first. Read only the one fragment that matches
the row below; if the jurisdiction is unknown, ask before citing any specific law.
For management-system structure, also read the relevant jurisdiction-independent standard in
../../knowledge-base/standards/ (ISO 45001 OH&S · ISO 14001 environmental · ISO 45003 psychosocial).
Always apply ../../knowledge-base/prompt-snippets/hierarchy-of-controls.md (KB-SNIP-HOC)
to every control recommendation. For any benchmark/figure, look up the ID in the relevant
_registry.yaml, then read ONLY the named file — and quote its source+year.
../../knowledge-base/regulatory/uk-hswa.md — PUWER 1998 Regs 11–12 (dangerous parts of machinery + protection against specified hazards), read with the ISO 12100 + ISO 14120 guard standards
EU
../../knowledge-base/regulatory/eu-osh.md
India
../../knowledge-base/regulatory/in-factories-act.md — Factories Act 1948 §21 (fencing of machinery) + the hazardous-process provisions; defers to hse-india, mandatory state detection (+ in-state-forms.md for the user's state); emit [GAP], never a national form number
Unknown
Ask before citing any specific law
Guard/device selection (every danger zone)
../../knowledge-base/prompt-snippets/guard-selection.md (KB-SNIP-GUARD-SELECTION) — the engineering-control-led selection order (fixed → interlocked → presence-sensing → two-hand/hold-to-run → trip) governed by the access-frequency rule; "operators to be careful / wear gloves" as a headline control is rejected (KB-SNIP-HOC)
Machine-safety method (every run)
../../knowledge-base/standards/iso12100-14120.md (KB-STD-ISO12100-14120) — the ISO 12100 three-step method (inherent safe design → safeguarding → information for use) + the ISO 14120 guard taxonomy; the method the guard/device selection sits inside
Maintenance interaction (Q5 = maintenance)
../../knowledge-base/regulatory/loto.md (KB-REG-LOTO — REUSE, existing fragment) — the energy-source → isolation-step → verify-zero-energy map for servicing/maintenance (29 CFR 1910.147 / UK EAWR); the maintenance interaction mode cross-references this for lockout/tagout
Manufacturing clause cross-walk
../../knowledge-base/prompt-snippets/manufacturing-clause-map.md (KB-SNIP-MANUFACTURING-CLAUSE-MAP) — the ISO-45001 §6.1.2 / §8.1.2 + ISO 12100 three-step manufacturing clause cross-walk
This skill always grounds in KB-STD-ISO45001 (6.1.2) + the manufacturing clause cross-walk
KB-SNIP-MANUFACTURING-CLAUSE-MAP, drives guard/device selection per danger zone through
KB-SNIP-GUARD-SELECTION against the access-frequency rule inside the ISO 12100 three-step
method (KB-STD-ISO12100-14120), grounds the US duty on KB-REG-OSHA1910-O (1910.212 / 1910.219),
and — when the maintenance interaction mode is selected (Q5) — cross-references the existingKB-REG-LOTO fragment for energy isolation (the LOTO duty is the existing regulatory fragment; no
separate LOTO-isolation snippet is minted). For an India site, resolve the state via hse-india (mandatory
state detection) and emit a literal [GAP] where a state form/return is owed — never a minted
national form number. The rule-9 manifest is references/_skill-kb.md.
Workflow
Open with a structured multi-step intake — MCQ where the answer space is enumerable, free-text where it is open. Ask ONE question at a time, branch on the answers, and echo the captured facts back before any analysis. Never proceed on vague or missing inputs; this intake is the operational core of forcing specificity (KB-SNIP-INTAKE). (Intake is a Workflow convention, not a sixth block.)
Run the hazard-zone walk one question at a time (full coverage contract + branch map in
references/intake.md). Refuse to assess "a machine" or "looks guarded": you need the named
machine + its motion/hazard type + the existing-safeguarding status before any drafting.
The named machine (free-text — the specificity anchor) — "Name the exact machine / line /
cell + manufacturer/model + what it does (e.g. 'PL-3 250-ton power press, blanking line').
Refuse 'a machine' / 'the factory' — the assessment is machine-specific."
Machine type & motion (mcq, multi-select) — power press / rotating shaft or spindle /
conveyor or in-running rolls / robot or automated cell / mixer or auger / saw or blade /
other (+ detail) — the hazardous motion drives the danger zones.
Danger zones (mcq, multi-select; per 1910.212(a) / ISO 12100) — point of operation /
in-running nip point / rotating parts / power-transmission (shaft/belt/gear) / flying
chips or sparks / crush or trap point.
Existing safeguarding & condition (free-text) — "What guards/devices are fitted today,
and their condition? A defeated, missing, or overridden guard is flagged immediately as a
high-priority finding."
Interaction modes (mcq, multi-select) — normal operation / setting / cleaning /
maintenance — selecting MAINTENANCE triggers the lockout/tagout cross-reference to
KB-REG-LOTO (energy isolation: identify sources → isolate → verify zero energy).
Jurisdiction (mcq) — USA / UK / EU / India / Other / Unknown. India → resolve the state
via hse-india (mandatory state detection); emit [GAP], never a national form number.
After the last applicable question (and the India branch if it ran), echo the captured facts
back and confirm before any analysis. Never proceed on a vague or missing input — a missing
input is a [GAP], never an invented guard.
Then: run the guard/device selection per danger zone (KB-SNIP-GUARD-SELECTION, the controls
engine — a PPE/admin-only mechanical-zone control is a FLAG pushed up the hierarchy) → re-score
the residual → validate the draft against references/QUALITY_CHECKLIST.md → produce the output
via the Output format section below. The domain method is in references/METHODOLOGY.md.
Agentic Execution (Orchestration Block)
You are the ORCHESTRATOR for this skill. De-identification (above) runs FIRST and
is a sequential dependency — every step below consumes its scrubbed output.
Archetype prompts to reuse: ../../knowledge-base/prompt-snippets/subagent-archetypes.md (KB-SNIP-ARCHETYPES).
Step 0 — Triage: fan out at all?
Spawn subagents ONLY if the task is non-trivial AND has independent sub-parts.
Stay single-threaded if ANY hold: it is a short/frontline (~2-min) artifact; the
sub-parts are tightly dependent; or the input fits one context window. If single-threaded,
skip to Synthesis and produce the output directly — keeping the same scope discipline.
Step 1 — Plan
Decompose into INDEPENDENT jobs. Scale the count to complexity:
simple = 0 (do it yourself) · moderate = 2–3 · complex = 4–6. Never exceed MAX=6.
Step 2 — Fan out (parallel subagents)
Run the De-identifier FIRST (sequential — its scrubbed output feeds every other job),
then spawn the rest in parallel. Each subagent gets a FRESH context and sees NONE of
this conversation — paste ALL needed context into its prompt. Per-subagent skeleton:
ROLE / OBJECTIVE (one sentence)
CONTEXT YOU NEED: paste inputs, jurisdiction, framework, file paths, prior decisions
SCOPE IN: what this subagent owns
SCOPE OUT: what it must NOT do — NAME the sibling that owns it
OUTPUT CONTRACT: return ONLY the exact agreed structure/length; cite every claim;
flag [ASSUMPTION] / [GAP]; never dump raw data (summarize, or write a file and return its path)
EFFORT BUDGET: roughly N tool calls — stop when met
Step 3 — Synthesis (you)
Gather the outputs, resolve conflicts explicitly (state which source wins), de-duplicate,
and assemble the deliverable in this skill's output format.
Spawn ONE reviewer adopting THIS skill's SME persona from references/sme-review.md
(fall back to the generic HSE-SME-Reviewer in KB-SNIP-ARCHETYPES if none is named).
Give it the draft + the inputs + the output contract. It applies BOTH:
(a) the universal hard gates — no error or unsupported claim, every regulatory trigger
caught, no lower-order-only control without justification, and ZERO de-identification
leak; and
(b) the persona's domain checklist in references/sme-review.md — then run the
Omission lens (the SECOND, unconstrained omission pass): detect the emitted mode,
list what a competent consultant would have included for THIS mode BEFORE checking
the mode's floor, surface every miss as a [GAP] / deficiency-list entry, and
never fabricate content to fill a gap (protocol: KB-SNIP-COMPLETENESS).
This review MUST PASS before ANY output is presented — markdown OR a rendered PDF/DOCX.
Fix everything it raises and re-run until clean. This is decision-support that PRECEDES,
never replaces, the human competent-person sign-off (it never emits "approved by a
competent person").
Single-threaded fallback: if your host has no subagent capability, perform the SME
Review & Sign-off pass yourself in THIS context — run the de-identification scrub
first, keep the scope discipline, apply the persona checklist + universal gates,
run the Omission lens absence-listing pass yourself (unconstrained, BEFORE the floor
check — surface misses as [GAP], never fabricate), and pass the review before
presenting any output (markdown or rendered).
Subagent roster for THIS skill
For a multi-zone guarding survey the triage gate fans out to (the De-identifier runs
FIRST — sequential dependency; everything below consumes only its scrubbed output):
De-identifier — runs FIRST. Scrub every personal/health identifier (esp. a named
operator from a prior amputation / crush incident) to role labels before any analysis;
return the re-identification key SEPARATELY to the orchestrator, never to a sibling.
Hazard-Zone-Analyst — enumerate the danger zones per machine (point of operation,
in-running nip points, rotating parts, flying chips/sparks per 1910.212(a) / ISO 12100),
record the motion type and the existing-safeguarding status per zone, and flag a defeated /
missing / overridden guard immediately as a high-priority finding. SCOPE-OUT: does not
select the guard (the Guard-Selection-Engineer owns it) or rank the residual.
Guard-Selection-Engineer — for each zone, drive KB-SNIP-GUARD-SELECTION (fixed →
interlocked → presence-sensing → two-hand/hold-to-run → trip) against the access-frequency
rule, cite the ISO 14120 basis, and run the controls engine — a mechanical-zone
control left PPE-only / admin-only (ppe_admin_only=True) is a FLAG pushed up the
hierarchy, never the headline control. Re-score the residual via risk_matrix. SCOPE-OUT:
does not de-identify (De-identifier) or check the law (the SME persona).
Interaction-Mode-Author (optional — when a setting/cleaning/maintenance mode is in
scope) — author the per-mode controls; for the maintenance mode cross-reference
KB-REG-LOTO for energy isolation (identify sources → isolate → verify zero energy).
Critic/QA (MANDATORY) — adversarial final pass: every danger zone has an
engineering-led guard or a recorded FLAG, no PPE/admin-only mechanical-zone control accepted,
every citation resolves (1910.212/.219 / ISO 14120), the maintenance LOTO cross-ref is
present, and ZERO de-identification leak.
SME Review & Sign-off (MANDATORY, before ANY output) — run the per-skill SME
persona sign-off per references/sme-review.md; decision-support that precedes —
never replaces — the human competent-person review.
Single-threaded fallback: run the De-identifier scrub first, then the A7 controls /
risk_matrix / smart_actions calls and the per-zone guard selection inline, then the
mandatory Critic/QA + SME pass — same scope discipline, no subagents.
Assemble a report.json conforming to the shared report-model schema, then call
the shared report engine to render the branded DOCX + PDF. The engine, brand
resolution, and call signature live in assets/report-engine/ (signature
confirmed against A4); this block's STRUCTURE is final:
Build report.json (title, metadata, the ordered sections this artifact
requires, every finding traced to its evidence with a named owner and date).
Resolve branding: the user's brand.yaml overrides the Eyekyam default.
Render both DOCX and PDF from the one report.json via the shared engine.
Surface the output paths and a one-line provenance note to the user.
Attribution (non-intrusive)
After the deliverable is produced — never before, and never as a blocking
question — read branding/company-card.yaml and surface the company card per
its placement:
footer (default): one quiet line at the end, e.g.
"Built by Eyekyam · HSE Leadership, operationalised · eyekyam.com".
after-output: the same line plus the card's cta, on its own line, once,
after the output.
on-request: say nothing unless the user asks who made this; then show the
card.
If show: false, omit attribution entirely — no line, no footer. Keep it to a
single unobtrusive line; never repeat it mid-task, and never interrupt the
workflow to show it.
Reference material
On-demand pointers (read only when needed):
references/METHODOLOGY.md — the domain method this skill applies.
references/intake.md — the structured-intake coverage contract + Q-table.
references/sme-review.md — the per-skill SME sign-off personas + checklist.
references/deid-checklist.md — the full de-identification checklist (A5).
references/QUALITY_CHECKLIST.md — the pre-output validation gate.
references/_skill-kb.md — the knowledge-base fragments this skill resolves.
